Development of the ability to read and to process complex spoken language is critical for academic success. Yet many individuals have low reading and oral language skills and thus may be poorly equipped to meet the challenges of our complex society. The existing literature suggests that the growth of reading and later language may be affected by early language and pre-reading skills, and by home environments. Indeed, it has been suggested that early disparities among children will remain or even increase over the school years (The Matthew Effect). However, the issue remains to be systematically explored, since studies of school achievement in relation to early experience have been lacking. We have recently carried out a longitudinal study of language growth in a diverse sample of children from 14 to 58 months. We propose to study these children over the school-age years. The proposed research is unique in its examination of language growth and reading in children from a wide variety of backgrounds whose early language and pre-reading skills have been examined in detail. Aim 1. We propose to extend our study of language growth by evaluating children's language over a longer period of language growth, examining their narrative skills as well as vocabulary and complex syntax. We will explore the patterns of development from 14 months until 10 years. Data from the first five year study showed sizeable individual differences and a substantial relation to caregiver input. In the next period of the grant, we will examine individual differences in oral language levels in the school-age years, exploring possible discontinuities in growth rates, reflecting, for example, growth of reading comprehension. The results of the proposed study should provide a more reasoned basis for decisions as to when and how to intervene to increase language skills of the population. Aim 2. We propose to examine how early language, pre-reading skills, and home environments are implicated in school learning. Preschool growth measures will be used as predictors, allowing us to distinguish among children with similar scores at one time point, but different patterns of growth. This is important because different input histories may have different implications for later learning. Outcome measures involve both reading and oral language. The data will allow us to systematically examine if discrepancies among children increase over the school years, as might be expected if children with better skills at the time of school entry learn more at school. Alternatively, since higher-performing children show greater growth over the summer than lower-performing children, this summer difference alone could explain the Matthew Effect. In order to pinpoint the sources of differences among children, we examine growth separately over the school year and summer. Investigation of predictors of individual differences should contribute to our understanding of ways to overcome the effects of poor early environments.